Investigation and Evaluation of Genetic Diversity of Plasmodium falciparum Kelch 13 Polymorphisms Imported From Southeast Asia and Africa in Southern China

Objectives: In this study, we aimed to analyse the genetic diversity Kelch 13 (K13) propeller allele of the Plasmodium falciparum isolates mainly imported from Southeast Asia and Africa in southern China, including the provinces of Yunnan and Guangxi. Methods: At enrolment, we collected blood samples from patients with confirmed cases of malaria infection between January 2012 and December 2017, for analysis. Individual patient information was obtained via a malaria surveillance system. The malaria infections and P. falciparum K13 mutations were diagnosed by using a nested polymerase chain reaction (PCR) method. Results: The K13 mutations were identified in 283 P. falciparum isolates from 18 counties in Yunnan and 22 counties in Guangxi. Of Forty-six isolates (46/283, 16.3%) that harbored K13 mutant alleles were detected: 26.8% in Yunnan (33/123) and 8.1% in Guangxi (13/160). A total of 18 different K13 mutations were detected. Only the F446I mutation was detected in Yunnan isolates, and F446I was more frequent (20/46, 43.5%) than other alleles. Further, the temporal distribution of the F446I mutation ratio from 2012 to 2015 exhibited no significant difference in Yunnan Province (2012, 2/13, 15.4%; 2013, 7/40, 17.5%; 2014, 7/33, 21.2%; 2015, 4/37, 10.8%, p = 0.121). A578S allele was the main K13 mutation (5/283, 1.8%) from Africa. The K13 mutants were present in 33.3% of indigenous isolates, 27.4% of isolates from Southeast Asia, and 7.9% of isolates from Africa. The analysis of 10 neutral microsatellite loci of 60 isolates showed that at the TAA109 locus, the expected heterozygosity of F446I (He = 0.112 ± 0.007) was much lower than that of wild type and other mutation types in Myanmar isolates. With respect to geographic distribution, TAA109 also exhibited a significant difference between isolates from Southeast Asia (He = 0.139 ± 0.012) and those from Africa (He = 0.603 ± 0.044). Conclusions: The present findings on the geographic diversity of K13 mutant alleles in P. falciparum may provide a basis for routine molecular surveillance and risk assessment, to monitor artemisinin resistance (ART) in China. Our results will be helpful for enriching the artemisinin resistance database in China during the elimination and post-elimination phases.


INTRODUCTION
Artemisinin-based combination therapies (ACTs) have been recommended as first-line drugs for the treatment of Plasmodium falciparum malaria infection in endemic regions by the World Health Organization (WHO). ACTs have contributed to a sharp reduction of the malaria burden in many countries (1). However, artemisinin resistance (ART), initially detected near the Thai-Cambodia border has spread across the Greater Mekong Subregion (GMS) and been confirmed in five countries (Myanmar, Cambodia, Laos, Vietnam, and Thailand) of the GMS (2,3). Therefore, the WHO have launched the Global Plan for Artemisinin Resistance Containment to contain ART-resistant isolates (4). ART and its derivatives have been widely used in China for nearly three decades (5). For example, Yunnan Province has a long history of ART monotherapy, and in vitro studies have shown reduced susceptibility of P. falciparum from the region to ART (6). It would be catastrophic if ART-resistant isolates have spread across the GMS to the border areas of China, especially to Yunnan and Guangxi provinces in southern China (7,8). Such spread is of great concern for malaria elimination in China. Furthermore, P. falciparum has recently exhibited a sharp increase in China, particularly in Yunnan and Guangxi provinces (9). Besides, Yunnan is one of the most severe endemic regions in China and most cases of P. falciparum malaria are imported from Southeast Asia (10).
The PfKelch 13 propeller (K13) has been identified as a molecular marker for ART-resistant isolates. More than 5% of surveyed patients carrying K13 resistance-confirmed mutations, and all of whom have been found to have either persistent parasitaemia by microscopy on day 3 or a half-life of the parasite clearance slope ≥ 5 h after treatment are considered as confirmed endemic artemsinin resistance (11). A total of 186 K13 alleles have been identified and five of them (C580Y, R539T, Y493H, N458Y, and I543T) have been confirmed as ART-resistant isolates (11). Furthermore, delayed clearance following ART treatment has been detected around the China-Myanmar border, with F446I identified as the predominant K13 mutant allele (7,12). In central China, no K13 resistance alleles have been validated, except in Henan Province where R539T and P574L have been identified as the main alleles found among migrant workers returning from African countries (13,14). Recently, a patient who returned from Equatorial Guinea to Jiangsu Province was identified as harboring M579I, which has been linked to ART resistance with a 2.29% in vitro survival rate by ring-stage survival assay (15). Those findings provide basic molecular surveillance data regarding the potential emergence of ART-resistant isolates in African countries. However, little is known about differences in the distribution and genetic diversity of the K13 propeller allele between locally acquired and imported samples, such as from Southeast Asia and Africa. Therefore, understanding the attributes of the K13 allele in ART-resistant isolates and its genetic diversity in China is of great importance, especially for areas bordering Myanmar and those with many malaria cases imported from African countries, to determine whether ART resistance is emerging or has already spread to China.
In this study, we aimed to investigate and evaluate the genetic diversity of the K13 propeller allele in Yunnan and Guangxi provinces in southern China, which have the greatest number of malaria cases imported from Southeast Asia and Africa, respectively. We sought to reveal the K13 mutation status and monitor the potential emergence of ART resistance in China.

Study Design
In this study, an indigenous case refers to malaria acquired by mosquito transmission in China. The imported case refers to the patient who acquired the illness from a known malaria-prevalent area outside China (16).

Study Sites and Samples
A total of 285 P. falciparum-infected blood samples were examined in this study. The individual data were obtained during 2012-2017 via a malaria surveillance system of the National Institute of Parasitic Diseases, Chinese Center for Diseases Control and Prevention. The study was conducted in Yunnan Province and Guangxi Zhuang Autonomous Region. Yunnan has a population of 47 million people living in its 16 prefectures and 129 counties. Among them, 18 counties bordering Myanmar were included in this study. Because no cases of P. falciparum infection were reported in Gongshan and Fugong counties; therefore, Lianghe and Jiangcheng counties were included instead because these are adjacent to border counties. For each year from 2012 to 2017, the annual reported malaria incidence rate in Yunnan Province was 0.18, 0.12, 0.11, 0.13, 0.07, and 0.05 per 10,000 residents, respectively. Guangxi is a province with some of the highest numbers of imported cases reported in China. The last indigenous case in Guangxi was reported in 2012; all subsequent cases have been imported from abroad, particularly from African countries. In 2013, owing to clusters of migrant workers returning from Ghana to Shanglin County in Guangxi, the number of imported malaria cases rose substantially, contributing greatly to the number of reported malaria cases nationwide in that year (17).
At enrolment, we collected blood samples from patients with a confirmed malaria diagnosis by microscopy, polymerase chain reaction (PCR) (18), or rapid diagnostic tests (Malaria Pf/pan, Wondfo Biotech Co., LTD., Guangzhou, China) in the laboratory in Yunnan and Guangxi provinces, from January 2012 to December 2017. Approximately 200 µL of finger-prick blood was spotted onto Whatman 3MM filter paper (GE Healthcare, Boston, MA, USA) and air dried. Each sample was labeled with the study number, name, and date and stored at −20 • C until DNA extraction.
All individual cases included in this study from the webbased reporting system (China Information System for Diseases Control and Prevention) were carefully reviewed and analyzed.

Preparation of DNA Template, Nested PCR, and Sequencing
Parasite genomic DNA from all blood-spot samples collected into microcentrifuge tubes and one filter spot (diameter 5 mm) was extracted using a QIAamp DNA Blood Mini Kit (QIAGEN, Valencia, CA), 500 µL elution buffer was added into each tube by following the dried blood spots protocol for the kit. The K13 propeller gene (PF3D7_1343700) was amplified using a nested PCR method as described previously (2). PCR products were separated and visualized using 2% agarose gel electrophoresis and GelRed (Shanghai Aberlong Biological Co., Ltd., Shanghai, China). The size of K13 fragment was 849 bp covering mutation sites from 427 amino acids to 709 amino acids. Sequences were analyzed using the BLAST website (http:// blast.ncbi.nlm.nih.gov/). Multiple nucleotide sequence alignment and analysis were performed using the DNAMAN sequence alignment editor (https://www.lynnon.com/pc/framepc.html).

Population Genetic Analysis
Population genetics were preferably assessed only in samples with single infections because the use of samples with multiple infections could result in bias. To measure the genetic diversity, the number of haplotypes (h), number of different alleles (N a ), number of effective alleles (N e ), number of private alleles (N p ), and the expected heterozygosity (H e ) were evaluated using the GenAlEx 6 microsatellite plugin (22). H e was calculated using the following formula: , and 2(n−1)/n 3 {2(n−2)[ (p 3 i -( p 2 i ) 2 ]} for sampling variance, where n is the number of sampled infections and p i is the frequency of the ith allele.

Statistical Analyses
Data were analyzed using Microsoft Excel 2003 and SAS version 9.2 (SAS Institute Inc., Cary, NC, USA). The map was created by ArcGIS 10.1 (Environmental Systems Research Institute, Inc., Redlands, CA, USA). The Fisher's exact test was used to assess the differences in gene polymorphisms between indigenous cases and imported cases from abroad, as well as the temporal distribution of mutant alleles. Calculated p-values were considered statistically significant with p < 0.05.

Ethical Considerations
The study was reviewed and approved by the ethical committee of the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention (NIPD, China CDC). This study was carried out in accordance with the recommendations of Chinese National Antimalarial Resistance Surveillance Plan, China CDC. The protocol was approved by the NIPD, China CDC. All subjects gave written informed consent in accordance with the Declaration of Helsinki.

Study Samples
A total of 285 isolates were collected: 123 isolates were collected from 18 counties in Yunnan, including 6 indigenous isolates and 117 imported isolates from Southeast Asia (116 isolates from Myanmar and 1 isolate from Laos). Another 162 isolates were collected from Guangxi, all of which were imported from African countries. Because 2 isolates were unable to be sequenced, therefore 283 effective isolates were finally obtained for analysis (Figure 1). All the 283 isolates were confirmed as positive by using three methods (microscopy plus RDTs plus PCR), the PCR gel picture were showed in Figure S1.
A total of 18 different K13 mutant alleles were observed, including 15 non-synonymous mutation sites and 3 synonymous mutation sites (L440L, C469C, and Y500Y) (Figure 2). F446I was the more frequently detected allele (20/46, 43.5%) than other mutation alleles. A total 18 isolates harboring the F446I allele were found in patients returning from Myanmar and 2 were found in patients from China (p < 0.05). Additionally, N458Y, which is associated with ring-stage survival assay, was found in two samples from Myanmar. The mutant allele A578S was detected in 6 patients, 5 from Ghana and 1 from Myanmar (p < 0.05). Further, three isolates harboring the A676D allele were all found in patients returning from Myanmar (

Microsatellite Analysis
We genotyped 60 isolates (including 17 from African countries and 43 from Southeast Asia) using analysis of 10 neutral microsatellite loci. According to the K13 single nucleotide polymorphism (SNP) sequencing data and sample size, and because F446I is the prominent mutant allele in the China-Myanmar border area, the microsatellite loci data were classified into three types: wild type (n = 16), F446I allele type (n = 20), and other mutant allele types (MT, n = 24). Generally, no significant difference in allele frequency was observed between wild type and MT; the H e value of F446I was lower than that of wild type and MT (Table S2). Since most of the F446I were imported from Myanmar, herein we found the F446I type at TAA109 locus (H e = 0.112 ± 0.007) had fewer alleles than wild type (H e = 0.385 ± 0.028) and MT (H e = 0.478 ± 0.036) in Myanmar isolates (Table 4). With respect to geographic distribution, TAA109 exhibited a significant difference between    isolates from Southeast Asia (H e = 0.139 ± 0.012) and those from African countries (H e = 0.603 ± 0.044) ( Table 5).

DISCUSSION
There are two great challenges facing the elimination of malaria in China. One is the presence of cross-border malaria along the China-Myanmar border in Yunnan Province. Even though local transmission has been interrupted, with no indigenous cases occurring in 2017, reintroduction may occur as the Anopheles mosquito remains present in the region (9). The other challenge is the increasing number of imported cases, with migrant populations arriving frequently from endemic countries or regions, such as Africa, which makes case management and surveillance very difficult. It is known that migrant populations are at risk of exposure to malaria and these groups are considered to carry and spread resistant parasites (23). In this study, we selected Yunnan and Guangxi, the two provinces in China with the highest number of malaria cases imported from Southeast Asia and Africa, respectively, to understand the genetic diversity of K13 allele polymorphisms and conduct population genetic analysis using microsatellite loci genotyping. Our findings provide evidence that areas near the border between China and northern Myanmar have a high prevalence of the F446I mutation in the K13 gene, which is similar to other previous reports (7,(24)(25)(26). To date, F446I has mostly  (12); however, this is not relevant to treatment outcomes in the Myanmar-India border area (27). In addition, A676D was previously identified in three studies including samples from northern Myanmar (26,28,29). N458Y is associated with ring-stage survival assay values >1% and has been proved to be a molecular marker associated  In general, there was a relatively low prevalence of K13 mutations among samples from Africa in this study. A578S had a higher prevalence than other mutant alleles in the African samples. This mutation has been reported as having no correlation with ART-resistant isolates in previous studies and is a K13 polymorphism commonly detected throughout Africa, suggesting that A578S possibly has a distinct origin and evolution pathway (32). However, the A578S mutation was detected in an isolate from one of our study participants who had returned from Myitkyina, a city located in northern Myanmar. S522I and A675V were found in only 1 isolate from Sierra Leone and Ghana, respectively, which may reveal that these mutations originated in Africa (33,34). Apart from this, the samples from Africa exhibited a high prevalence of synonymous mutation sites; except for Y500Y, three had been previously reported (35), and the other two mutations were newly detected.
In K13 population genetic analysis, we conducted a microsatellite assay to assess and evaluate geographic evolution. F446I was seen to exhibit a lower H e value than wild type and MT alleles from Yunnan Province and northern Myanmar, suggesting that this mutant allele evolved locally in the region (Table S2). Thus, F446I could be selected and adopted for use as a molecular marker in the China-Myanmar border region, to identify the geographic origins and dissemination of ART-resistant P. falciparum. However, further in vivo and in vitro studies are needed, such as clinical efficacy surveillance associated with F446I mutations. Interestingly, the TAA109 locus, showed varying H e values among F446I, wild type, and MT, as well as diversity between isolates from Southeast Asia and Africa. This highlights the need for further research to understand the relationship between TAA109 and the F446I mutation, such as genome-wide sequencing and linkage disequilibrium analysis. We propose that ART resistance is unlike resistance to chloroquine and sulfadoxine-pyrimethamine in that the latter are associated with relatively fixed mutations in codons. For example, isolates from Africa exhibit the highly resistant pfcrt (C 72 V 73 I 74 E 75 T 76 ) and dhfr (P. falciparum dihydrofolate reductase gene, C 50 I 51 R 59 N 108 I 164 ) alleles (36,37). The mechanism of K13 candidates associated with ART-resistant P. falciparum could be a multigenetic process and could also be combined with other molecular markers, such as pfmdr1, pfmdr2, pfatpse 6, and pfarps 10, all of which have been previously reported to be related to ART resistance (38)(39)(40). The next work will carry out msp1 and msp2 PCR genotyping of samples used in microsatellite analysis, to explore the further reason of the low diversity at TAA109 which may due to the selective sweep at K13 locus.
For the limitation, of the malaria cases studied, only six cases were indigenous malaria. Considering that we intended to analyse the diversity of locally acquired infections and potential evolution of ART resistance in China, the number of samples were not adequate.
In summary, the present data indicated a high prevalence of the F446I mutant allele, detected among both indigenous and imported malaria cases from Myanmar, especially for counties along the China-Myanmar border. In addition, a distribution pattern of K13 mutant alleles was observed that differed from other regions of the GMS or Africa. Microsatellite analysis indicated that the F446I allele had a much lower H e value than both wild type and MT at locus TAA109. As for imported cases from Africa, few K13 mutant alleles related to ART-resistance were found, with A578S predominating in African samples. Routine molecular surveillance and risk assessment should be conducted to monitor ART resistance throughout the study region. The present data will be helpful in development of a genetic molecular marker to trace malaria infection sources during the malaria elimination and post-elimination phases, to achieve the goal of eliminating ART resistance in P. falciparum throughout the GMS.

AUTHOR CONTRIBUTIONS
JF designed and conducted the study, and drafted the article. XK and HT carried out the statistical analysis. HY carried out the molecular experiment. DX carried out the epidemiology survey and recording. HZ and KL supported the study and performed sample collection.